'Real World' Photoshop Benchmark - Speed Test

[Leafminer]

I guess barefeats was right about mixing capacities of RAM modules, I might have to try and sell my factory modules and buy some 2GB modules from OWC instead.

Has anyone else experienced similar results?

Not too sure about that Jeb...

25 seconds

MacPro 2.8 GHz 8core (2008)
6 GB RAM (2 Apple OEM+4 GB (2x2) OWC)
Nvidia 8800GT
CS 3
Mac OS X (10.5.2)
OS / Program drive: 500 GB WD stock
Scratch: 75 GB partition of Samsung 750GB F1
Fresh restart


Mark

 

[jeznav]

Did the test 3 times and I got 24 seconds all the time.

Results...

Test 1: 24.58
Test 2: 24.26
Test 3: 24.75

 

[dukebound85]

31s

quad 2.4ghz
3gig ram
80 gig 5400rpm laptop drive

 

[snagitseven]

2.8 Dual Quad Mac Pro 8800GT Video Card
8 GB RAM
CS3 running on 1TB F1 Samsung HD w/system, apps
Scratch disk in 141GB partition on 2nd internal 1 TB F1 Samsung (Time Machine uses balance of this HD)
3 apps open
30 History states, 5 Cache levels, PS set to 84%, No restart (was too lazy)

45 seconds

 

[snagitseven]

2.8 Dual Quad Mac Pro 8800GT Video Card
8 GB RAM
CS3 running on 1TB F1 Samsung HD w/system, apps
Scratch disk in 141GB partition on 2nd internal 1 TB F1 Samsung (Time Machine uses balance of this HD)
3 apps open
30 History states, 5 Cache levels, PS set to 84%, No restart (was too lazy)

45 seconds

Ok. I followed directions: reset history to 1, cache to 4, 100% usage with restart, no other apps open.

29 seconds.

 

[YahonMaizosz]

Hi guys..
I am a PC user and had built my own PC.. I had taken the speed test and would like to post my result here, just for reference. I hope that this helps..

1st try (No reboot after 18 hours of usage): 18.30 seconds

Here are my specs:

Q6600 @ 3.6Ghz
3 x 150GB RAPTOR RAID 0
8GB OCZ Platinum Edition DDR2 960
ASUS Maximus Formula
GeForce 7950GX2

This thread had been a great source of information regarding the photoshop speed test. I hope that it continues this way.

EDIT: After a restart, the result is 17 seconds.

 

[OSXconvert]

MacPro speeds Tiger vs. Leopard?

Has anybody tested the difference on the test with PS CS3 on 10.4.11 vs. 10.5.2 on a MacPro with at least 4gb of RAM? I'm curious if Leopard is faster. Back when 10.5 came out, I compared it with 10.4.11, and Leopard was actually a tad slower.

 

[amik]

26.3 seconds

CS3
10.5.2
Early 2008 2.8ghz 8-core
10gb ram (4x2 from transintl + 2x1 stock sticks)
3*750gb Samsung F1s, first has osx and apps, second has music & videos + scratch, and the third has photos. Stock 320gb barracuda not in use at the moment.
8800gt

 

[mperkins37]

I'm expecting CS3 any day now, I'll let you know.

 

[graphicjoe]

This is a first post. I am a PC person who is planning to switch to a Mac Pro. However, ths test has made me have second thoughts.

I ran the test eight times on my Dell 380. It has a 955 EE CPU, 4 GIGs of RAM, four 500 GIG SATA HDs, and an NVIDIA 1400 video card. On each run my time was either 49 or 50 seconds, with only two 50 second runs.

I found that the memory setting did not affect the times.

Joe Stephenson

 

[YahonMaizosz]

I am a PC user too. What are you planning to use ur Mac Pro for? As far as I know (If you consider the hardware only) Mac Pro is good for multi threaded applications since it has 8 cores and basically, it is the cheapest workstation in the market, even against you building your own workstation!

 

[adrianr]

This is a first post. I am a PC person who is planning to switch to a Mac Pro. However, ths test has made me have second thoughts.

I ran the test eight times on my Dell 380. It has a 955 EE CPU, 4 GIGs of RAM, four 500 GIG SATA HDs, and an NVIDIA 1400 video card. On each run my time was either 49 or 50 seconds, with only two 50 second runs.

I found that the memory setting did not affect the times.

Joe Stephenson

To be honest it seems a lot of the power of the new Mac Pros is completely wasted in Photoshop. Not to mention the FB-DIMM ram which for smaller photo tasks will be slower. If you intend on editing huge files or batchprocessing massive amounts of stuff, it may be worthwhile. Otherwise, spec up a PC for half the price and enjoy a faster photo processing machine.

 

[Leafminer]

This is a first post. I am a PC person who is planning to switch to a Mac Pro. However, ths test has made me have second thoughts.

I ran the test eight times on my Dell 380. It has a 955 EE CPU, 4 GIGs of RAM, four 500 GIG SATA HDs, and an NVIDIA 1400 video card. On each run my time was either 49 or 50 seconds, with only two 50 second runs.

I found that the memory setting did not affect the times.

Joe Stephenson

Why second thoughts? The new MacPros are running 25-29 seconds in this test.

 

[contoursvt]

The person already has a machine that does it in say double the time...and thats a pretty intensive test. So what may take 2 seconds on his PC, it will take say 1 second on the Mac Pro (just an example of day to day tasks), so really, it may not be worth spending $3000 for this.

Not saying its not worth it for the Mac Pro, but the numbers are meaningless unless you're doing some kind of task which takes a very long time to complete. Even for me as a PC guy, going from a dual core to a quad really got me very little tangable improvement even though benchmarks improved lots.

Why second thoughts? The new MacPros are running 25-29 seconds in this test.

 

[adrianr]

Even for me as a PC guy, going from a dual core to a quad really got me very little tangable improvement even though benchmarks improved lots.

Unfortunately this is true for most. Too much software doesn't make good enough use of 4 cores let alone 8. A Mac pro would benefit those who use multithreaded video encoding software (or similar) best. Something that can really use all the power of the machine.

Photoshop, put simply, doesn't.. And Unless you're working on something that *needs* more than 8Gb of ram, even then you're probabaly going to be alright with a Raid0 scratchdisk array. You could probabaly lean to live with the slight performance drop in that scenario for how little money it's cost you.

 

[graphicjoe]

I have given more thought to changing to a MP. I well realize that the additional speed is a relatively small matter for most everyday PS tasks, even though I work with large files and do much panoramic stitching, something that demands a lot of processing power. During a good part of generating a large panorama consisting of 5-7 30 Meg files, the performance monitor shows that all four threads are maxed out, and most of the machine's 4 gigs of RAM are in use.

I mind that Adobe is rumored to be coming out with an update that will make use of quad cores on dual processors. That may change the equation a bit. It is also important the realize that there is some point at which more machine power will make scant difference in actual throughput of work. As it is, most of our computers spend a lot more time waiting for us than we spend waiting for them.

But as you all know, raw speed is only a part of what makes for a desirable computer, and not always the biggest part. The OSX is a well evolved OS and the limited experience I've had using MAC's made me think they were very reliable and stable. I took a two week work shop at Anderson Ranch Art Center where about a dozen Power MAC's were used day and part of the night for two weeks without a reboot, hang, or anything from the machine that caused a work stoppage. That really impressed me. I was sometimes frustrated when attempting basic tasks that are easy for me in Windows because I know Windows very well, having used it daily since 3.1. I realize that it would not take me long to learn how to do everything I need to do on a MAC, but still, that is a small factor to consider.

I am in the process of selling one house and buying another, so I am putting the MAC decision on hold until we are settled in our new home. In the meantime I may buy some additional RAM for my 380, taking it to 8 gigs. That way, PS can have all it wants and there will be plenty of memory left for Windows and other applications.

Finally, thank you for the responses to my post,

Joe Stephenson

 

[gunbu]

First post

Hey guys,
This is my first post! So I ran the test last night on my "bottom rung" Mac Pro 2.8 Quad. I used all the settings requested in the docs, and I got 35 seconds before restart, and 29 seconds afterwards.

 

[trcwest]

New 2008 MBP

I got 41 Seconds with all the settings set right. Does this sound right for my specs??

Compared to the retouch artists specs page it does "
MisterNet Macbook Pro Intel C2D 2.33 2GB 120GB 5400rpm CS3b,48s" thats what one user posted

My specs 2.5ghz c2d, 4gb Mem,250 gb 5200rpm HD,
On all the right settings and CS3.

IS there and more people with the new MBP 2008 just to compare to see if mines about right!!

Thanks for replying

 

[MarkCombs]

31 seconds

31 Seconds
MacPro (8 Core / January 2008)
Upgraded 640 Gig WD (new 2 platter drive - very fast)
2 gigs of ram

I'm going to run this again when I have 4 gigs of ram installed which will take me to 6 gigs...

Mark

 

[YahonMaizosz]

I don't understand why my Q6600 is faster with your 8-core Mac Pro?? Isn't having more processor going to make everything run faster?

Btw, I get 17 seconds after restart.

 

[eXan]

I don't understand why my Q6600 is faster with your 8-core Mac Pro?? Isn't having more processor going to make everything run faster?

Btw, I get 17 seconds after restart.

You probably have a faster hard drive than Mac Pro.

And no, having extra processor means nothing if the software cant utilize it (Photoshop is one of those apps). If you didnt notice, the 4 core Mac Pros are just as fast in this test as 8 cores.

 

[contoursvt]

Maybe its because you've overclocked it to 3.6Ghz?

I don't understand why my Q6600 is faster with your 8-core Mac Pro?? Isn't having more processor going to make everything run faster?

Btw, I get 17 seconds after restart.

 

[YahonMaizosz]

You probably have a faster hard drive than Mac Pro.

And no, having extra processor means nothing if the software cant utilize it (Photoshop is one of those apps). If you didnt notice, the 4 core Mac Pros are just as fast in this test as 8 cores.

After going through the last 10 pages of the Thread I just realized that the 4 core and 8 core Mac Pro only has a little difference despite having 4 more processors. I guess, Photoshop is unable to utilize all the cores. What a pity.

 

[adrianr]

I don't understand why my Q6600 is faster with your 8-core Mac Pro?? Isn't having more processor going to make everything run faster?

Btw, I get 17 seconds after restart.

It is most likely due to the RAM rather than the CPU or Hardrive. Well, the CPU will attribute to it a little (Probabaly more due to the Q6600 being optimized for desktop apps, and the Xeons being server chips optimized for such tasks), but as this test doesn't utilize all 4 cores at 100%, it wouldn't make much difference.

Mac Pros use Fully Buffered Dimms. These are typically slower than un-buffered DDR2 found in most desktop PC's. As this test is essentially a test of RAM (Which photoshop pretty much is in general untill you start hitting the scratchdisk alot) the PC's faster ram will allow it to complete faster.

 

[Blaina]

I minute 40 seconds

MacBook pro 2.33 Core 2 Duo /3G RAM/ Hitachi 200G/7200 harddrive/ Leopard 10.5.2/ CS2